Carbon-Aware Microservice Deployment for Optimal User Experience on a Budget

2025-06-26

AI Summary

This work addresses the joint optimization of carbon budget compliance and low-latency service availability in microservice deployment within data centers. We propose the first holistic framework for hourly carbon-intensity-aware microservice version selection and horizontal autoscaling. Methodologically, we formulate an integer linear programming model that integrates real-time carbon intensity signals, dynamic workload fluctuations, and hard QoS constraints (latency and availability), enabling carbon-aware scheduling at the microservice level in real time. Our key contributions are: (i) extending carbon-aware scheduling to the fine-grained, highly dynamic microservice layer for the first time; and (ii) introducing a version-aware autoscaling mechanism that supports heterogeneous deployment of multiple service versions. Experiments demonstrate that our approach strictly satisfies hourly carbon budgets across diverse application configurations, improves average user satisfaction by 18.7%, increases revenue by 12.3%, and responds to minute-scale variations in both carbon intensity and workload.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Hardware-aware ML
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The carbon footprint of data centers has recently become a critical concern. So far, most carbon-aware strategies have focused on leveraging the flexibility of scheduling decisions for batch processing by shifting the time and location of workload executions. However, such approaches cannot be applied to service-oriented cloud applications, since they have to be reachable at every point in time and often at low latencies. We propose a carbon-aware approach for operating microservices under hourly carbon budgets. By choosing the most appropriate version and horizontal scaleout for each microservice, our strategy maximizes user experience and revenue while staying within budget constraints. Experiments across various application configurations and carbon budgets demonstrate that the approach adapts properly to changing workloads and carbon intensities.

Problem

• Optimizing microservice deployment for carbon efficiency
• Balancing user experience with carbon budget constraints
• Adapting to dynamic workloads and carbon intensities

Innovation

• Carbon-aware microservice deployment strategy
• Optimal version and scaleout selection
• Adapts to workload and carbon changes